Prognostic value of biological aging in TAVR for aortic stenosis: a combined prospective and retrospective cohort study
R Tang, H Li, W Li, Y Li, Y Yao, X Yang, M ChenAbstract
Background
Biological aging (BA) is a comprehensive marker reflecting systemic physiological aging, offering critical context for individual variability based on chronological age. A series of studies have demonstrated that BA is associated with the prognosis of cardiovascular diseases and outcomes following percutaneous coronary intervention. However, the predictive ability of various BA indicators calculated from clinical biochemistry biomarkers for outcomes in aortic stenosis (AS) patients after transcatheter aortic replacement (TAVR) remains unknown. This study aimed to To evaluate the role of BA indicators for prognosis of AS patients post-TAVR.
Methods
This study included a retrospective cohort and a multicenter prospective cohort who underwent TAVR for AS. We estimated biological age using the Klemera–Doubal method (KDM) and homeostatic dysregulation (HD) method, both involving biomarker selection through correlation analysis, machine learning based feature selection, and explainability analysis. And we calculated PhenoAge and ProBNPAge using well-established algorithms. Multivariable-adjusted Cox proportional hazards models incorporating restricted cubic splines were applied to evaluate the association between accelerated biological age and adverse clinical outcomes.
Results
A total of 937 patients (mean age 72.6±7.8 years) and 470 patients (mean age 71.6±7.3 years) were included. Five biomarkers were selected for HD and KDM calculation. In the retrospective cohort, a 1-SD increase in HD, log(HD) and PhenoAge, and accelerated aging, showed positive associations with 3-year mortality and MACE (HR > 1.0; p < 0.05) in four adjusted models. In the prospective cohort, HD was associated with higher 2-year mortality and readmission (HR > 1.5; p < 0.05) in four adjusted models. Restricted cubic splines demonstrated a linear dose-response relationship between BAs and adverse outcomes (p for non-linearity > 0.05).
Conclusion
Biological age indicators particularly HD, demonstrate high prognostic utility for predicting post-TAVR mortality and adverse events. Incorporating these comprehensive markers provides a systemic physiological aging perspective, enhancing risk stratification and treatment selection for aortic stenosis patients.